If there's any way you can provide us with a testcase, that would help enormously. I'd be happy to take a copy of the app, in confidence of course - I promise to nuke it as soon as I can derive a testcase!

I'm hoping LZOs will be fully working in swf9/10 soon - then you could send us a binary library... Thanks!

Regards,
Max Carlson
OpenLaszlo.org

On 1/8/10 5:15 PM, Chris Kohlhardt wrote:
http://jira.openlaszlo.org/jira/browse/LPP-8697

On Fri, Jan 8, 2010 at 4:59 PM, Max Carlson <[email protected]
<mailto:[email protected]>> wrote:

    Chris,

    Any chance you could file a bug at http://jira.openlaszlo.org/ and
    attach the screenshots/testcase there?  If there's a regression in
    swf9, we really want to take care of it!

    Regards,
    Max Carlson
    OpenLaszlo.org


    On 1/8/10 4:19 PM, Chris Kohlhardt wrote:

        The following message bounced when I tried to send screenshots
        of the
        problem.

        On Fri, Jan 8, 2010 at 4:17 PM, Chris Kohlhardt
        <[email protected] <mailto:[email protected]>
        <mailto:[email protected] <mailto:[email protected]>>> wrote:

            I compiled our using the nightly LPS build, and this results
        in our
            application looking very jumbled.  SWF9 and SWF10 both show
        the same
            issue.

            If I turn the debugger on, the application looks correct.
              (screenshots attached)

            It sort of looks like constraints aren't working as expected....
              but I don't have any evidence besides visual evidence to
        prove this.

            I spent some time trying to isolate the issue, but haven't
        had any
            luck so far.   Our application is pretty complicated, so
        it's pretty
            tough to isolate issues.

            Any ideas?

            -chris


            On Fri, Jan 8, 2010 at 3:43 PM, Henry Minsky
        <[email protected] <mailto:[email protected]>
        <mailto:[email protected] <mailto:[email protected]>>>
        wrote:

                Did you mean the issue is that your code (which you've been
                running in swf9) compiled for swf10 has some artifacts,
        or that
                compiling to swf9  in the nightly build has problems?



                On Fri, Jan 8, 2010 at 5:34 PM, Chris Kohlhardt
        <[email protected] <mailto:[email protected]>
        <mailto:[email protected] <mailto:[email protected]>>> wrote:

                    I just gave the nightly build a quick spin, and
        immediately
                    ran into rendering issues which I assume are related to
                    SWF9....  Is SWF9 support going away?

                    We have decided not to adopt SWF10 yet because we have
                    customers who are in the 'Enterprise' and the data
        we have
                    suggests Flash 10 adoption is still far less than
        90% there.
                      I think the Adobe numbers are misleading

          
(http://www.adobe.com/products/player_census/flashplayer/enterprise_penetration.html)
                    and the analytics on our web site suggest Flash 10
        has maybe
                    80% penetration.

                    -chris

                    On Fri, Jan 8, 2010 at 11:04 AM, Henry Minsky
        <[email protected] <mailto:[email protected]>
        <mailto:[email protected] <mailto:[email protected]>>>
        wrote:

                        We just made some changes to significantly
        reduce the
                        RAM required for SWF9/10 compiles. You can try
        them out in
                        a nightly build, and tell us if you see any
        improvement
                        (or any new bugs, god forbid)

                        regarding the 'incremental compile' option, If you
                        compile from the command line, the incremental
        option
                        will be useless right now, since the
                        cache it stores is in RAM. If run on the server,
        I don't
                        know if it makes any difference either, it's really
                        just a placeholder feature now and does not have an
                        efficient implementation,  it requires more work
        to be
                        optimized to make much difference.



                        On Fri, Nov 20, 2009 at 4:39 PM, Chris Kohlhardt
        <[email protected] <mailto:[email protected]>
        <mailto:[email protected] <mailto:[email protected]>>> wrote:

                            After a good amount of work, we've managed
        to get
                            our application completely migrated to
        OL4.6.1 and
                            SWF9.

                            Thank you very much to everyone involved in
        making
                            the SWF9 runtime a reality.  The performance of
                            Gliffy is so much faster now, it's almost
                            unbelievable.  We're entering QA next week,
        and we
                            expect to release SWF9 Gliffy in mid December.

                            One thing we noticed is that compilation of
        SWF9 is
                            a lot slower.  After some digging, we were
        able to
                            speed things up by:
                            - setting compiler.swf9.incremental=true in
                            lps.properties
                            - allocating at least 2GB of memory to the
        tomcat
                            instance running the lps
                            - moving developers to a pure 64bit OS
        (Clint moved
                            to Windows 7 after a long stint with XP)

                            Are there any other performance tips to
        consider?

                            thx!

                            -chris




                        --
                        Henry Minsky
                        Software Architect
        [email protected] <mailto:[email protected]>
        <mailto:[email protected]
        <mailto:[email protected]>>






                --
                Henry Minsky
                Software Architect
        [email protected] <mailto:[email protected]>
        <mailto:[email protected]
        <mailto:[email protected]>>





Reply via email to